The Literature

In the field of agricultural economics, comprehensive assessments of corn markets, ethanol markets, and agricultural subsidies are readily found. Some of the more pertinent studies are outlined in the following sections to provide a basis for the present paper. The discussion of prior research and literature will be subdivided into three categories: agricultural subsidies, ethanol markets, and corn markets.

2.1 Agricultural Subsidies
Over the past ten years total agricultural subsidies in the United States ranged between $12 and $36 billion per year with an average of just over $23 billion per year (US OMB 58-60). The first agricultural subsidy program was the 1862 Morrill Act, which established the land-grant colleges. During the 1930s, subsidies began to take hold because of the Agricultural Adjustment Act and the New Deal, as did commodity price supports and production controls, marketing orders to limit competition, import barriers, and crop insurance (Edwards 2008). Many changes have been made to these programs over the years, but the fundamental central planning aspects have not. Uncertainty in the production of some agricultural commodities, frequently related to the unpredictability of nature or to human-generated interruptions either of supplies of imported food or of inputs to domestic production, is often the reasoning given for subsidy programs (Legg 2003).

        Each time reforms to these programs are presented to Congress they are either rejected or result in an increase, rather than a decrease, in subsidy payments. Currently, there are eight major types of farm subsidies: direct payments, marketing loans, countercyclical payments, conservation subsidies, insurance, disaster aid, export subsidies, and agricultural research and statistics. Wilfred Legg (2003), a prominent figure on agricultural and environmental policy issues and the current Head of Policies and Environment Division in the OECD Agricultural Directorate, goes into great detail in his presidential address, outlining both the definition and the measurement of agricultural subsidies¹. Generally speaking, subsidies inherently create winners and losers; sectoral policies coupled with electoral cycles build a domestic constituency that supports the continuation among the few winners who gain a lot but rarely the losers who each lose a little (Legg 2003). Agricultural subsidies have become an important part of corn farmers’ production decisions and therefore must be included in any discussion of the economics of agricultural markets.

2.2 The Ethanol Market
Ethanol has seen explosive growth in popularity and production in recent years. Bio-based fuels such as ethanol provide potential solutions to urban air quality, global warming, and excessive dependence on imported oil, as well an economic solution to high crude oil costs (Ferris and Joshi 2004). Policy incentives exist at both federal and state levels that further support the growth of renewable fuels. An example of these is the Clean Air Act of 1990, which imposed mandatory oxygenate levels upon gasoline in areas with air quality issues. Also, the Energy Policy Act of 2005 introduced the Renewable Fuel Standard, which requires US fuel production to include a minimum amount of renewable fuels each year; the program starts at four billion gallons in 2006 and reaches a mandatory minimum of 7.5 billion gallons in 2012 (Tokgoz and Elobeid 2008). One way to meet these regulations and to reduce emissions is to blend gasoline with ethanol, accounting for about 79% of the US oxygenate supply in 2006 (Energy Information Administration 2008). In the US, the most common ethanol blend with gasoline is 10% (E-10), which can be used in any standard unleaded vehicle. Ethanol is also produced in an 85% blend (E-85) that can only be used in flex-fuel vehicles (FFVs), which run on gasoline, ethanol, or any combination of the two.

        Ferris and Joshi (2004) used an empirical model to examine the determinants of increased ethanol production and its subsequent impact on the agricultural industry. To determine these impacts, the authors proposed five events that could contribute to increased ethanol production, and then examined their cumulative effects in different combinations. These events are as follows:

1. Fourteen state or Federal ban on MTBE, a substitute to ethanol in gasoline blends.
2. Congress passing the Federal Renewable Fuel Standard.
3. Increased use of ethanol as a blending agent due to high gasoline prices that tend to make ethanol a cost effective octane enhancer.
4. Supreme Court ruling to enact revised national air quality standards for 8-hour ozone concentrations.
5. United States Department of Agriculture (USDA)’s Commodity Credit Corporation (CCC) providing incentive programs for bioenergy production.

       Through the use of a multi-sector econometric model called AGMOD that contains over 400 equations and more than 700 variables, the authors produced quantitative results concerning both ethanol and corn markets. Ferris and Joshi (2004) assumed that corn will be used as feedstock for ethanol production, resulting in ethanol production projections for 2010 that ranged from 3,250 million gallons to 4,670 million gallons. The proportion of total corn production used in ethanol production ranged from 9.5% to 14.8% in 2010. The price received by corn producers increased nearly 30% from 2003 to 2010 ². The final conclusions made by Ferris and Joshi (2004) posited:

1. Corn ethanol demand is likely to increase rapidly due to proposed changes in energy and environmental policies.
2. Agricultural commodity prices will increase more sharply in the short run, followed by moderate increases due to expanded acreage under grain production.
3. Increased use of ethanol fuel is likely to be beneficial to farmers, improve air quality and contribute to energy security by marginally reducing dependence on foreign oil.

These results point towards an increasing use of ethanol as a blending agent with gasoline. This increased use would in turn have ripple effects in the agricultural sector, specifically to corn.

       Another study on ethanol and its effects on agriculture, by Tokgoz and Elobeid (2006), modeled the link between ethanol, energy, and crop markets. The authors systematically decompose the factors affecting the international ethanol market and then proceed to contrast the US and Brazilian ethanol markets. One fundamental difference between the two markets is the type of inputs used. In Brazil the major feedstock for ethanol production is sugarcane, as opposed to corn in the US. Both feedstocks face competition in the intermediate input market. In the US, ethanol competes with livestock industries that use corn as feed. In Brazil, however, sugarcane used to produce ethanol could otherwise be used in the production of sugar rather than ethanol. Tokgoz and Elobeid (2006) examine the use of ethanol as a substitute for gasoline and also as a complement in the production of gasoline.

        The results provided by Tokoz and Elobeid (2006) showed that a 20% increase in gasoline prices in the US would result in a 4% decline in composite gasoline consumption. At the same time the share of fuel ethanol in composite gasoline consumption increased by 2.5% due to substitution. The total ethanol consumption, after the increase of gasoline prices, declined by 1.5% because the increase use of ethanol as a blending agent was less than the decrease use of blended fuel. The net effect on corn demand predicted by the authors was a 0.6% increase in consumption. These results were built upon the assumption that the number of FFVs in the US is limited in the short run. In the long run, with an increase in the use of flex-fuel vehicles, substitution of ethanol for gasoline increases and thus higher gasoline prices will lead to increased ethanol consumption. Conclusions reached in the study depended upon the composition of the domestic vehicle fleet. The proportion of FFVs determines whether ethanol acts as a substitute for or complement of gasoline.

2.3 The Corn Market
Several papers addressing estimates of corn demand will be used to construct the model utilized in this research. One of the more recent papers regarding corn markets that will be used to construct the hypotheses of the current study is that by John Marsh (2007). He examined the farm-level relationships that exist among the corn, livestock, and poultry markets by assessing interdependencies on inputs, demands, and supplies of each commodity. The paper developed an econometric model that integrates four sectors through mutual dependency of structural demands and supplies. These factors are used to estimate cross-sector impacts of changes in corn loan rates, corn export demand, and fertilizer costs on the demand for and supply of livestock and poultry, or inversely, the effect of livestock and poultry meat demand on the demand and supply of feed corn.
            The focus upon farm-level production reflects different degrees of vertical integration/coordination in the sectors. For example, the feeder pig market is not separated due to an industry dominated by integrated farrow-to-finish operations. The sectors defined in the model are: feeder cattle, slaughter cattle, slaughter hogs, wholesale broilers, and corn production. The supply and demand curves are theoretically based on first-order necessary conditions of firm profit maximization. First principles of the optimization problem give input demands as a function of own input prices, substitute input prices, output prices, and technology. Output supply functions depend upon own prices, substitute prices in production, input prices, and technology. For the corn sector, Marsh presented the following equations, which are summarized in Table 2.3.1:

Table 2.3.1 Marsh’s Supply and Demand Variables

 
The econometric model is based upon autoregressive distributed lags, in which agricultural supply would be a function of expected output and input prices, with expectations formed by parameter weights on lagged output and input price variables; that is, future values are dependent through weights on lagged values. Three-stage least squares is used to estimate the model. 

            Many factors could be introduced as determinants of demand for corn. Prior research by the Economics Research Service (ERS), a subsidiary of the USDA, on the price determination of corn and wheat provides another basis for the assumptions made in this research. Hoffman and Westcott (2008) created a price determination model for corn and wheat. The authors found the significant factors in the supply of corn to be beginning stocks, imports, and production. In the same manner the significant factors for corn demand are food, seed, and industrial use, feed and residual, and exports. These factors are then broken down into the relevant economic variables to be used in the price determination model.

            Hoffman and Westcott recognized the important role that government programs play in the formation of the equilibrium price and quantity of corn. The most significant of these programs was the price support and commodity storage programs. Through price support programs, farmers receive a loan from the government at a designated loan rate per unit of production while raising their crops as collateral. Farmer-owned-reserve programs provided storage subsidies to farmers to store grain under loan for three to five years. Farmers who had grains in this program would not be able to sell their grain unless the price rose above a preset level. The manner in which the model was built and the significant factors developed provides much of the intuition used in the formulation of this research.

            Mathew Holt (1992) produced another influential paper on the estimation of corn demand. He used a multimarket bounded price variation model under rational expectations. Price supports were directly incorporated through the market clearing mechanism and price expectations in the supply functions. Holt used restricted reduced-form price equations with conditional expectations. Demand for corn was a function of the price of corn and soybeans, the price of livestock, exports, and a time trend. The supply of corn was a function of expected soybean production, the expectation of the effective producer price of corn, seasonal growing conditions, and a binary variable to discount the effects of the 1983 payment-in-kind program and the severe drought.

1The Organization for Economic and Co-Operation and Development in Paris, France brings together the governments of countries committed to democracy and the market economy from around the world to support and propagate economic growth.

2 Since the paper (Ferris and Joshi 2004) was written in 2004, ethanol production has far surpassed their predictions and has in fact reached over 4,800 million gallons in 2006, reaching 14.3% of total US corn supply. For more information see Iowa Corn Growers Association <http://www.iowacorn.org/cornuse/cornuse_3.html> (Accessed 25 Nov. 2007). 

Methodology